[Top] [All Lists]

RE: Consensus? #733 Outsourcing principle

Subject: RE: Consensus? #733 Outsourcing principle
From: John C Klensin
Date: Thu, 13 Jan 2005 19:04:18 -0500

--On Thursday, 13 January, 2005 21:21 +0100 "Wijnen, Bert
(Bert)" <bwijnen@xxxxxxxxxx> wrote:

> Whether you call it RFP or RFI (sorry I don't do these things,
> so I may be mis-using terminology), the result is (I think)
> that  if bidder A says they can do it with 2, Bidder B with 5
> and Bidder C with 15 people, then I Think one would find the
> number for C to be bloated (for whatever reasons).
> Anyway... enough about this as far as I am concerned

Bert, let me take a last shot at explaining why the problem
isn't just about terminology (and why you may have missed the
key problem) by changing your scenario a little bit.    

(1) Suppose one puts out a document soliciting proposals and
price quotes on some activity X (that falls into the range of
what is usually called an "RFP").  And let's assume you get
proposals offers back from five entities, with proposed head
counts of
     A  2
     B  3
     C  15
     D  16
     E  18

Now, one hypothesis is that C is bloated and D and E are even
more bloated.   Another hypothesis is that A and B fundamentally
don't understand the problem or, even worse, one of them is
clueless and the other has gamed the situation and concluded
that, by the time the IAOC figures out that the job cannot be
done with 2 or 3 people, it will be too expensive to recompete
the deal so they will have no choice by to renegotiate the
contract to cover C's headcount at E's price.

Note that (other than the fact that I moved the A and B numbers
closer together for clarity), the scenario above cannot be
distinguished from a variation of your scenario in which D and E
read the RFP, concluded that it was poorly-written enough to
attract clueless low bids, and decided it was worth making
proposals about, so all you would see would be the A, B, and C

This is where a proposer qualification process of some sort (RFI
or otherwise) comes in.  It functions at least to reduce the
odds of someone clueless making a proposal, but may not
eliminate fraudlent or "gamed" bids.  An even more complex
process may involve a dialogue with potential bidders (usually
RFI-driven), followed by generating a very specific RFP that
everyone agrees covers the work and collecting what then become
simple competitive bids (from among those who participated in
the drafting process) from the proposers.  That sometimes works
well, but is very complicated to administer and is rarely
appropriate for small contracts (if only because proposers can
rarely be persuaded to play).

(2) _However_, none of the helps with what is really the key
question, which is whether X needs doing.  Perhaps some much
smaller project, Y, would do the job instead.  And that was the
problem I was trying to comment about earlier today.   No RFP,
RFI, or equivalent process will help with the problem of whether
the IAD and IAOC makes good decisions about how to define the
tasks and scope that are optimal for the IETF.  Our protection
there is to make it clear that optimality and efficiency are
important, that we put high value on defining tasks to require
the minimal headcount and money needed to get them done well,
that we make sure those decisions are visible to the community,
and that the IAOC is held accountable for them.  IMO, those
principles should be shown clearly in the BCP.   Anything else
-- outsourcing versus hiring, RFPs versus RFIs, and so on-- are
all issues that the IAOC should work out (and review regularly)
but are distractions (or handcuffs) if they appear in the BCP.


Ietf mailing list

<Prev in Thread] Current Thread [Next in Thread>